Toxicity and drug resistance limit the efficacy of long term antiviral therapy for cytomegalovirus (CMV) disease in severely immunosuppressed hosts, including those failing anti-HIV therapy and in transplant recipients who develop primary CMV infection. The long-term objective of this grant is to characterize the viral mutations that confer resistance to CMV drugs in clinical use. We now have a reasonable understanding of the mutations in the CMV UL97 kinase and UL54 DNA polymerase genes that confer resistance to the licensed drugs (ganciclovir, foscarnet and cidofovir), yet many new mutations continue to be identified. Available information is widely used for genotypic diagnosis of CMV resistance and selection of alternate therapies. A promising experimental CMV UL97 kinase inhibitor drug maribavir appears to have high potency, oral bioavailability and low toxicity in early clinical trials, and is currently undergoing Phase III trials in transplant populations. Genetic mechanisms of maribavir resistance are being studied, with mutations already identified in UL97, UL27 and elsewhere. Maribavir resistance mutations outside of UL97 are unexplained, but probably compensate for shutoff of the UL97 kinase. Further analysis should clarify the role of UL97 in viral replication and as an antiviral drug target. The specific aims for the upcoming period are (1) determine the effects on drug resistance and viral growth of single and multiple CMV mutations observed in clinical specimens, (2) genotypic and phenotypic characterization of maribavir resistance, and (3) define the genetic mechanisms of viral adaptation to UL97 inhibition. Experimental priorities include continued improvements in recombinant phenotyping technology, validation of additional resistance mutations including those for maribavir, assessment of cross-resistance, identifying adaptive mutations outside UL97 and determining the interaction with UL97 and cellular kinases.